CIntX: A Software Tool for Calculating the Intrinsic Exchange … · 2013. 9. 6. · are herein...

CIntX: A Software Tool for Calculating the Intrinsic Exchange Rates of Labile Protons in Proteins

T. Richa and T. Sivaraman* Structural Biology Laboratory,

Department of Bioinformatics, School of Chemical and Biotechnology, SASTRA University,

Thanjavur – 613401, Tamil Nadu, India.

Abstract Estimation of extrinsic (kex) and intrinsic (krc) exchange rate constants for labile protons of proteins is indispensable to determine the residue-specific free energies. While kex of labile protons in proteins are determined by experimental methods, the krc of those protons are calculated based on the parameters derived from the model compound studies, in general. We, herein, describe a computational tool, CIntX, which calculates the krc of labile protons in proteins at defined experimental conditions such as pH, temperature, solvents and ionic strength using three-dimensional structures of the proteins. The program also accounts structural parameters (amino acid sequence, cis-trans proline conformations, disulfide bonds and ionisable groups in the proteins) influencing the krc of labile protons in a fully automated manner. The robustness of the tool on predicting the krc of labile protons in proteins has been validated using data reported in the literature. The program is publicly available at http://sblab.sastra.edu/cintx.html Keywords: Algorithm, Computational tool, Disulfide bonds, Free energy, H/D exchange, Intrinsic Exchange rates, Ionic strength, Proline conformation and Protection factor.

1. INTRODUCTION

The structure-function relationships of proteins have intrigued the researchers for many decades [1]. Each protein adopts a unique three-dimensional structure with defined stability, which is important for its biological activities. The stability of proteins is calculated by measuring the free energy differences between the folded (N) and the unfolded conformations (U) of the proteins at ambient conditions [2,3]. Hydrogen-deuterium (H/D) exchange put forward by Linderstram-Lang is a powerful method to estimate free energy of exchange (ΔGHX) of proteins at residue level [4]. In a typical H/D exchange experiment, when a protein is dissolved in deuterium oxide (D2O), labile amide protons (NHs) of the protein begin to exchange with solvent deuterium and the exchange phenomena is represented by the following two-state model equation proposed by Hvdit [5,6]. Closed (NH) Open (NH) Exchanged. (1) In this model, the kop and kcl are rate constants for the unfolding and the folding reactions of proteins under native conditions, respectively. Exchange takes place only from the unfolded state of proteins with the rate constant of krc, which can be predicted based upon the parameters derived from studies on the model compounds [7]. In this context, it should be mentioned that SPHERE, a program developed by Roder, calculates krc values of backbone amide protons (NHs) of proteins from their primary structures [8]. In this paper, we are herein describing a novel computational tool, CIntX, which calculates krc values of labile protons located in the backbone, side-chains and termini of proteins at defined experimental conditions from their tertiary structures. The program requires only four inputs: PDB file of the protein, Temperature in Kelvin, pH and Ionic strength. The CIntX, in

a fully automated fashion, accounts for the effect of structural parameters such as amino acid sequence, cis-trans proline conformations, disulfide bonds and ionisable groups in proteins on the estimation of krc values for labile protons in the proteins. The uniqueness of the tool on predicting the krc

values for the labile protons of proteins and its application on determining ΔGHX of the proteins have also been discussed in detail.

2. METHODS The CIntX algorithm has been implemented using PERL scripting language [9]. The program accepts both amino acid sequences (represented by single letter codes) and PDB (Protein Data Bank) co-ordinates of proteins as query file. In order to calculate the krc values, the program takes temperature in Kelvin, pH in pD (pD = pH+0.4), ionic strength in molarities and activation energies in cal/mol [7,10]. The values of krc and ΔGHX of labile protons are expressed in per minute and kcal/mol, respectively. While a PDB file is an input, the program automatically defines the cis/trans conformations of Xaa-Pro (Xaa can be any one of the twenty standard alpha amino acids and Pro denotes proline) peptide bonds and patterns of disulfide bonds in the protein using the following relationship: D = {(UX-PX) 2 + (UY-PY) 2 + (UZ-PZ) 2}1\2 (2) where D is the distance between the Cα of Xaa and Cα or Cδ of proline residue in angstrom; U and P stand for Xaa and Pro residues of Xaa-Pro peptide bond, respectively; X, Y and Z are the atomic co-ordinates of an atom considered. The Xaa-Pro peptide bond is considered to have trans-conformation, when the distance between Cα of Xaa and Cα of Pro (Cα —Cα) is greater than the distance between Cα of Xaa and Cδ of Pro (Cα—Cδ). Similarly, when Cα—Cδ > Cα—Cα, the Xaa-Pro peptide bond is considered to adopt cis-conformation [11]. In

kcl

kop krc

T. Richa et al /J. Pharm. Sci. & Res. Vol.4(6), 2012, 1852 - 1858

1852

order to determine the disulfide bond patterns in proteins, the distance between the sulphur atoms of thiol groups from two cysteins is calculated using the equation (2) and the two cysteines are considered to be disulfide bonded, if the calculated distance is within 2.3 Å. For each successful run, the program provides two sets of the krc values for labile protons in proteins (one for accounting the effect of ionisable side-chains and another one for not accounting the effect). The CIntX determines the fraction of protonated (FHA) and deprotonated (FA) forms of ionisable groups in proteins using equations (3) & (4), respectively and the equations were derived from Henderson–Hasselbalch equation (mathematical derivations are not shown). FHA = 1/(1+10x-y). (3) FA= 1- FHA. (4) where x and y denote the pD of protein solution and the pKR of ionisable groups, respectively. The program then recalculates the krc of NHs that are influenced by the ionisable groups (detailed description is given under the results and discussion heading). In addition, the program displays structural parameters such as Xaa-Pro conformations and disulfide bond patterns and derived parameters such as protection factors and residue-specific free energy values in kcal/mol provided NMR-derived kex values of labile protons are given as input parameters. A detailed list of instructions to use the tool is made available at http://sblab.sastra.edu/cintx.html

3. RESULTS AND DISCUSSION 3.1 Mathematical derivations of equations for calculating krc of labile protons in proteins The values of krc for labile protons of proteins can be estimated using the equation (5) as described by Bai et al. [7,10]. krc = ka * Ra* 10 (-pD) + kb *Rb * 10 (pD-pKD) + kw * Rw. (5) where ka, kb and kw are rate constants of acid, base and water catalyzed exchange reactions, respectively; Ra, Rb and Rw are the effect of residues that are on the left and right sides of amide protons at acidic, basic and neutral conditions, respectively; pKD is the molar ionization constant of D2O; pD

is the pH-meter reading corrected to deuterium effect. Thus,

the equation (5) accounts the effect of primary sequence and solution pH on the krc of labile protons under consideration. The effect of temperature on the intrinsic exchange rates of the labile protons is calculated using the following equation. k(T) = k(293) exp(-E[1/T -1/293]/R) . (6) where k(T) and k(293) are rate constants at desired temperature and 293 K, respectively; E is the activation energy and its value for acidic, basic and neutral exchange reactions are 14 kcal/mol/K, 17 kcal/mol/K and 19 kcal/mol/K, respectively; T is the absolute temperature in Kelvin and R is the gas constant. The effect of temperature on the acid catalyzed exchange rate constant is as shown in the equation (7). log k(acid)T = log k(acid)293K

+ (Ea /4.57) * (1/293 -1/T) (7) By combining the equation (7) and the component representing acid-catalyzed reaction in equation (5), the following equation (8) could be derived. log k(acid)T

= log ka + log Ra- pD + (Ea /4.57) * (1/293 -1/T). (8) Considering, X = log ka + log Ra + (Ea /4.57) * (1/293 -1/T), the equation (8) can be rewritten as shown either in the equation (9) or in the equation (10). log k(acid)T

= X- pD. (9) k(acid)T = 10(X-pD). (10) Similarly, temperature-corrected base catalyzed and water catalyzed equations could also be derived. Substituting all the three temperature-corrected parameters in the equation (5), the following equation (11) is obtained. Hence, the equation (11) calculates krc values of labile protons in proteins in a single step accounting for the effect of neighbouring amino acids, pH and temperature. krc = 10(X - pD) + 10(Y + pD - pKD) + 10(Z). (11) where, X = log ka + log AL + log AR + (Ea /4.57)*(1/293 -1/T). (12) Y = log kb + log BL + log BR + (Eb/4.57)*(1/293 -1/T). (13) Z = log kw + log BL + log BR + (Ew/4.57)*(1/293 -1/T). (14) In the above equations, AL and AR denote the effect of residues that are on the left and right sides of amide protons at acidic conditions, respectively. Similarly, BL and BR denote the effect of residues that are on the left and right sides of amide protons at basic conditions, respectively [7].

Table 1: Comparing the actual and the predicted (by CIntX program) conformations of Xaa-Pro peptide bonds in Cytochrome C, Cobrotoxin and RNase A.

Protein (PDB ID)#

No. of Prolines

Position of Prolines

Distance measured by manual method*

Actual Conformation of Xaa-Pro peptide

bond

Distance calculated by CIntX Predicted conformation of

Xaa-Pro peptide bond Cα-Cα Cα-Cδ Cα-Cα Cα-Cδ

1HRC 4

30 3.78 2.80 Trans 3.781 2.796 Trans 44 3.81 2.96 Trans 3.814 2.957 Trans 71 3.81 2.85 Trans 3.809 2.854 Trans 76 3.79 2.64 Trans 3.790 2.638 Trans

1COD 2 12 3.69 2.64 Trans 3.694 2.644 Trans 44 3.66 2.02 Trans 3.658 2.017 Trans

5RSA 4

42 3.86 2.90 Trans 3.858 2.898 Trans 93 3.04 3.88 Cis 3.039 3.883 Cis 114 2.91 3.82 Cis 2.912 3.817 Cis 117 3.83 2.88 Trans 3.818 2.878 Trans

#1HRC: Cytochrome C, 1COD: Cobrotoxin and 5RSA: Rnase A. *PyMol, molecular visualization tool [22], was used for determining the conformations of Xaa-Pro peptide bonds in the proteins manually. Distances are

shown in angstrom.


1853

Table 2: Comparing the actual and the predicted (by CIntX program) cysteine(s) and cystine(s) of Cytochrome C, Cobrotoxin and Rnase A.

Protein (PDB ID)

No. of Cys#

Position of Cys

Distance measured by manual method Actual

Conformation of Cys residue

Distance calculated by CIntX Predicted conformation of

Cys residue S-S Pair Distance S-S Pair Distance

1HRC 2 14

none 8.71 Cysteine none 8.712 Cysteine 17

1COD 8

3 C3-C24 1.40 Cystine C3-C24 1.404 Cystine

24


41


54


60

5RSA 8


84


95


110


72 #Cys denotes cysteine residue. *The manual distance measurements for determining the cysteine and cystine residues in the proteins were carried-out using PyMol molecular visualization

tool [22]. Distances are shown in angstrom.

Figure 1. Flowchart outlines key-steps of the CIntX program.


1854

Figure 2. Correlation between the krc values estimated by manual calculations and the CIntX program for the NHs in proteins (A) Coborotoxin

(1COD) and (B) Cytochrome C (1HRC) at pH 3.2, 298 K.

3.2 Determining prerequisite structural parameters for estimating krc of labile protons in proteins Proline residues are linked to its preceding amino acid in polypeptides through an imide bond instead of amide linkages by which other amino acids are connected. Amide bonds prefer negligible percentage of (about 0.03%) cis-conformations in the unfolded states, whereas imide bond (Xaa-Pro) prefers remarkable percentage of cis-conformations in the unfolded states and the percentage varies (6-38%) depending on the chemical properties of the residue (Xaa) preceding proline [12,13]. The Xaa-Pro bond isomerises between its cis and trans conformations in the unfolded states of proteins under native conditions. In order to accurately predict the H/D exchange behaviour of labile protons influenced by the Xaa-Pro peptide bonds, the program determines the conformations of Xaa-Pro peptide bonds in proteins using the equation (2). Table 1 validates the robustness of the program on predicting cis-trans proline conformations in proteins such as Cytochrome C (1HRC), Cobrotoxin (1COD) and Rnase A (5RSA). A quick inspection to the table suggests that the actual conformations and the CIntX predicted conformations of proline residues in the proteins are in perfect agreement. Similarly, the krc of NHs of cysteine and cystine residues in proteins remarkably differ from each other [7]. Moreover, the krc of residues that are succeeded by cysteine/cystine residues are also differentially influenced by them. In order to account the effect of these residues, CIntX determines the conformations of each cysteine residue in proteins. When the distance between the sulphur atoms of thiol groups from two cysteins is within 2.3 Å, the two cysteines are considered to be disulfide bonded [14]. The program determines the cysteine/cystine conformation by sampling all possible cysteine pairs in proteins. The distance between any two sulphur atoms is calculated using the equation (2) described in the method section. Table 2 shows robustness of the program on predicting the cysteine/cystine residues in Cytochrome C, Cobrotoxin and Rnase A.

3.3 Estimation of krc values of backbone labile protons in proteins from their three-dimensional (3D) structures The outline of the program is schematically represented in Fig.1.The program calculates the values of krc for labile protons in proteins with four inputs: PDB file of the protein, pH, Temperature, and Ionic strength. In outline, the program has four stages for each complete cycle. First, the program determines number of proline and cysteine residues and their conformations using the input PDB file, itself (discussed in the above section). Second, krc of labile protons are calculated using equation (11), which accounts the effect of pH, temperature, and primary structure of proteins on the krc. The effect of neighbouring residues on the NH is calculated in the program as shown herein. /* Calculation of krc for the NH of s3 of S */ S = (s1, s2, s3........sn); for (i=1; i<n; i++) Ra = As2 * As3; Rb = Bs2 * Bs3; krc = ka * Ra* 10(-pD) + kb *Rb * 10(pD-pKD) + kw * Rb

end for; In the above representations, ‘S’ denotes total number of residues in the given input polypeptide chain. Each amino acid in the polypeptide chain is denoted as‘s’ and its subscript defines the position of the amino acid in the chain. Residue at i-1th position and ith position are considered to be on right and left parts to the NH of ith residue, respectively. A and B are standard reference parameters of the amino acids at acidic and basic/neutral conditions, respectively [7,15]. Moreover, depending on the ionic strength (high > 0.5 M or low < 0.5 M) and solvents (H2O or D2O) defined, the program assigns the values for ka, kb and kw that are stored as default parameters in the program and calculates krc for NHs of proteins in the second stage. The krc values of NHs in Cobrotoxin (1COD) and Cytochrome C (1HRC) have been calculated using the program and the calculated data have been compared with the krc values of the NHs as determined by manual method (Fig. 2). The data in Figure 2A and 2B


1855

were fitted to a simple linear equation and the fitted parameters such as slope and positive correlation coefficient were found to be 0.996 and 0.999, respectively, for both of the proteins. These observations unambiguously demonstrate that the reliability of the CIntX on calculating the krc of NHs in proteins. The program, then, calculates residue-specific free energies and protection factors of proteins with the only requirement of having NMR-derived kex of NHs in the proteins. In the third stage, the program recalculates the krc of NHs by accounting the effect of ionisable groups in proteins

and calculates krc of labile protons that are not NHs in the fourth stage of the program (discussed in the next headings). 3.4 Revaluation of krc values of backbone labile protons

in proteins after accounting the effect of ionisable side chains

In general, polypeptide chains consist of about 25% amino acids having ionisable groups (COOH/NH3/SH/OH) in their side chains [16, 17]. These ionisable groups will be fully or partially protonated depending on their pKR under defined experimental conditions. Bai et al. have determined the effect of protonated and deprotonated forms of a few of

Table 3 :The percentage of protonated and deprotonated forms of ionisable groups in proteins at pD 3.6 and their dissociation constants (pKR).

Residues pKR % Protonated forms % Deprotonated forms Asparatate 3.67 54 46 Glutamate 4.25 18 82 Histidine 6.54 100 0 Cysteine 8.55 100 0 Tyrosine 9.6 100 0 Lysine 10.4 100 0

N-terminal 8.00* 100 0 C-terminal 3.67** 54 46

* pKR of N-terminal amino group. ** pKR of C-terminal carboxylic acid group.

Table 4: Values of krc and ∆GHX of NHs of a few amino acids in cobrotoxin at pD 3.6, 298 K have been tabulated, herein.

Sl. No. Residue# Uncorrected* Corrected** ∆∆GHX krc ∆GHX krc (min-1) ∆GHX

1 E2 0.13 2.26 0.10 2.08 0.182 C3 1.04 3.62 0.84 3.48 0.143 H4 2.18 NA@ 2.18 NA NA4 N5 2.52 NA 2.52 NA NA 5 E21 0.31 1.4 0.23 1.22 0.186 T22 0.25 NA 0.20 NA NA7 D31 0.97 NA 0.35 NA NA8 H32 3.01 NA 1.32 NA NA9 R33 0.97 NA 0.97 NA NA10 E38 0.33 3.57 0.24 3.38 0.1911 R39 0.36 2.93 0.29 2.8 0.1312 E51 0.12 2.19 0.09 2.01 0.1813 I52 0.06 2.23 0.05 2.11 0.1214 D58 0.93 NA 0.33 NA NA15 R59 0.58 3.46 0.26 2.98 0.4816 C-terminal$ 0.26 NA 0.07 NA NA

#Amino acids are denoted by single letter codes. *The krc values of NHs are calculated without accounting the effect of ionisable groups in the protein. **The krc values of NHs are calculated by accounting the effect of ionisable groups in the protein. @NA represents the residues for which the values of kex are not available. $C-terminal carboxylic acid group.

Table 5: Estimation of the krc values of labile protons in the side-chains of cobrotoxin at pD 3.6, 298K. Sl. No. Residue# Side Chain Proton Rate Constant krc (sec-1) Time Constant τ (sec)

1 N5 HE 1.47 0.682 Q6 HE 0.98 1.023 Q6 HZ 0.41 2.424 Q7 HE 0.71 1.405 Q7 HZ 0.30 3.326 Q10 HE 0.57 1.767 Q10 HZ 0.25 4.028 N23 HE 1.07 0.939 R28 NδH 0.17 5.97

10 W29 N1H 0.56 1.7911 R30 NδH 0.10 10.1312 R33 NδH 0.86 1.1713 R36 NδH 0.14 7.0114 R39 NδH 0.25 4.0115 N48 HE 1.53 0.6616 N53 HE 0.76 1.3217 R59 NδH 0.22 4.5018 N61 HE 0.99 1.0119 N62 HE 2.22 0.45

# Amino acids are denoted by single letter codes.


1856

these amino acids (Asp, Glu, and His) on the intrinsic exchange rates of labile protons [7]. It is usually assumed (in the calculations of exchange rates of NHs in proteins) that an ionisable groups will be fully protonated when its pKR is greater than the solution pD and fully deprotonated when its pKR is lower than the pD of the solution. However, when there is not much difference (< 1.0) between the pKR of an ionisable group and the solution pD, the ionisable group in a polypeptide chain will have significant amount of its protonated and deprotonated forms under the defined experimental conditions. Table 3 shows the pKR of ionisable groups of amino acids (Asp, Glu, His, Cys, Tyr, and Lys) and their percentage of protonated/deprotonated forms at pD 3.6, 298 K. The CIntX determines the fraction of protonated (FHA) and deprotonated (FA) forms of those ionisable groups using equation (3) and equation (4) (refer method section) and then recalculates the krc of NHs that are influenced by the ionisable groups using the equation (15) described below herein in the third stage of the program. krc = 10(X - pD) + 10(Y + pD - pKD) + 10(Z). (15) where, X = log ka + log (FHA*AL++ FA*AL-) + log (FHA*AR ++ FA*AR-) +

(Ea /4.57)*(1/293 -1/T). (16) Y = log kb + log (FHA*BL++ FA*BL-) + log (FHA*BR ++ FA*BR-) +

(Eb/4.57)*(1/293 -1/T). (17) Z = log kw + log (FHA*BL++ FA*BL-) + log (FHA*BR ++ FA*BR-) +

(Ew/4.57)*(1/293 -1/T). (18) wherein ‘+’ and ‘-’ signs in the subscripts of the standard reference parameters of a residue indicate the protonated and deprotonated forms of the corresponding amino acid in the polypeptide chain, respectively [7,15]. Table 4 shows the recalculated krc values for the NHs of cobrotoxin, which are influenced by ionisable groups present in the protein. Residue-specific free energy (ΔGHX) values of the NHs have also been calculated using NMR-derived kex values of the NHs at identical exchange conditions (pD 3.6, temperature 298K) of the protein [18]. The highest ΔΔGHX was found to be 0.48 kcal/mol for R59 of the protein at pD 3.6, 298 K, as the residue is preceded by D58, an amino acid having ionisable group. This observation reveals that discrepancy between the krc-corrected and the krc-uncorrected will be remarkable for NHs depending on their positions in the polypeptide chain and the exchange-conditions such as pH and temperature. In the fourth stage, the CIntX calculates krc values of labile protons present in the side chains of arginine, asparagine, glutamine and tryptophan using equation (11) in a single step. Table 5 shows krc values of labile protons from a few side chains of cobrotoxin at pD 3.6 and 298 K. The labile protons of these side-chains in the protein are exchanging in the time span of 0.5 – 10.1 seconds (Table 5). The krc values of side-chain labile protons will be powerful probes to determine the folding rate(s) of tertiary structural formations of proteins in which the labile protons are protected from the solvent exchange under native conditions.

Error-free estimation of krc values for labile protons of proteins is indispensable to estimate accurate ΔGHX (∆GHX = - RT ln (kex/krc)) of proteins. The exchange of free energy is important to understand the relationship between three-dimensional (3D) conformations and stabilities of proteins as each protein adopts a specific, well-defined 3D structure, which is important for its biological activities [19]. Moreover, ΔGHX analysis provides clues on understanding the mechanism of unfolding of proteins (two-state/multi-state processes) and on analysing the 3D structural architectures (domains organization) of proteins [20, 21]. Thus, it is obvious that estimation of an accurate ∆G (free energy change) for proteins at ambient conditions is indispensable to unambiguously address the thermodynamic and kinetic events of proteins. In these backgrounds, we strongly believe that CIntX, fully automated and user-friendly computational tool facilitated with web server (http://sblab.sastra.edu/cintx.html), is very useful for estimating krc values of all types of labile protons (backbone, side chains, N-terminal and C-terminal positions) of proteins.

4. CONCLUSION We have herein developed a software tool (CIntX) to calculate the intrinsic exchange rate constants (krc) of labile protons from backbone, side-chains and termini positions of proteins at defined experimental conditions. CIntX is a fully automated and user-friendly tool in its functions and requires four inputs to achieve the task: PDB file of protein, temperature in Kelvin, pH, and ionic strength. The program accounts the effect of Xaa-Pro peptide bonds, disulfide bonds and ionisable side chains on the krc estimations using the 3D structural co-ordinates of the given protein. The program also calculates residue-specific free energies and protection factors of proteins with the only requirement of having NMR-derived kex of NHs in the proteins. Estimation of precise ∆GHX (free energy of exchange) for proteins at ambient conditions is indispensable to unambiguously address the thermodynamic and kinetic events of the proteins. In these backgrounds, CIntX is a unique tool of this kind for calculating krc of labile protons in proteins. The web server of the program is publicly available at http://sblab.sastra.edu/cintx.html

REFERENCES [1] Dobson, C. M., Karplus, M., Curr. Opin. Struct. Biol. 1999, 9, 99 – 101. [2] Pace, C. N., Scholtz, C. M., in: Creigton, T. E. (Ed), Protein Structure - A

Practical Approach, Oxford University Press Inc., New York 1997, pp. 535 – 566.

[3] Brockwell, D. J., Smith, D. A., Radford, S. E., Curr. Opin. Struct. Biol. 2000, 10, 16 – 25.

[4] Englander, S. W., Mayne, L., Bai, Y., Sosnick, T. R. Protein Sci. 1997, 6, 1101 – 1109.

[5] Hvidt, A., Nielsen, S. O., Adv. Protein Chem. 1966, 21, 287 – 366. [6] Englander, S. W., Sosnick, T. R., Englander, J. J., Mayne, L., Bai, Y.,

Curr. Opin. Struct. Biol. 1996, 6, 18 – 23. [7] Bai, Y., Milne, J. S., Mayne, L., Englander, S. W., Proteins. 1993, 17, 75

– 86. [8] Zhang, Y., Protein and peptide structure and interactions studied by

hydrogen exchange and NM, Ph.D. Thesis, University of Pennsylvania 1995.


1857

[9] Wall, L., Christiansen, T., Orwant, J., Programming Perl, O'Reilly Media Inc., Sebastopol 2000.

[10] Connelly, G. P., Bai, Y., Jeng, M., Englander, S. W., Proteins. 1993, 17, 87 – 92.

[11] Wuthrich, K., NMR of proteins and nucleic acids, John Wiley & Sons., New York 1986.

[12] Reimer, U., Scherer, G., Drewello, M., Kruber, S., Schutkowski, M., Fischer, G., J. Mol. Biol. 1998, 279, 449 – 460.

[13] Huyghues –Despointes, B. M. P., Scholtz, J. M., Pace, C. N., Nat. Struct. Biol. 1999, 6, 910 – 912.

[14] Bhattacharyya, R., Pal, D., Chakrabarti, P., Protein. Eng. Des. Sel. 2004, 17, 795 – 808.

[15] Molday, R. S., Englander, S. W., Kallen, R. G., Biochemistry 1972, 11, 150 – 158.

[16] Grimsley, G. R., Scholtz, J. M., Pace, C. N., Protein Sci. 2009, 18, 247 – 251.

[17] Thurlkill, R. L., Grimsley, G. R., Scholtz, J. M., Pace, C. N., Protein Sci. 2006, 15, 1214 – 1218.

[18] Sivaraman, T., Understanding the folding and the unfolding of three – finger toxins from Taiwan cobra (Naja naja atra). Ph.D. Thesis, National Tsing Hua University, Taiwan, 1999.

[19] Huyghues-Despointes, B. M. P., Pace, C. N., Englander, S. W., Scholtz, J. M., in: Murphy, K. M. (Ed.), Protein Structure, stability and folding, Humana Press Inc., New Jersey 2001, pp. 69 – 92.

[20] Arrington, C. B., Teesch, L. M., Robertson, A. D., J. Mol. Biol. 1999, 285, 1265 – 1275.

[21] Krishna, M. M. G., Hoang, L., Lin, Y., Englander, S. W., Methods. 2004, 34, 51– 64.

[22] DeLano, W. L., The PyMOL Molecular Graphics System. DeLano Scientific, San Carlos, CA, USA 2002.


1858

Date post:	20-Feb-2021
Category:	Documents
Upload:	others
View:	2 times
Download:	0 times

CIntX: A Software Tool for Calculating the Intrinsic Exchange … · 2013. 9. 6. · are herein...

Documents